Fuzzy modelling of visual texture: coarseness, contrast and directionality properties

摘要

The analysis of the perceptual properties of texture plays a fundamental role in tasks like semantic description of images or content-based image retrieval using linguistic queries. In this paper, we propose to model these properties by means of fuzzy sets defined on the domain of some representative measures. In our approach, the membership functions associated to these fuzzy sets are obtained by learning a functional relationship between the computational values given by the measures and the human perception of the properties. The performance of each fuzzy set is analyzed and tested with human assessments, allowing us to identify the most suitable model to represent each property.